Driving Cluster-Level Trust for Artificial Intelligence in V2X Communications

The advent of autonomous and connected vehicles has put the spotlight on Vehicle-to-Everything (V2X) communication, which enables seamless data exchange between vehicles, infrastructure and external systems. While Artificial Intelligence (AI) plays a central role in improving decision-making and optimizing vehicle networks, the dynamic and uncertain nature of V2X environments makes it necessary to build trust in AI-driven systems. This paper presents a novel cluster-based approach to promote cluster-level trust within V2X communication. By utilising a belief function with cluster level confidence methodology, we provide a robust framework for dealing with uncertainty, merging evidence, and improving the reliability of AI-driven predictions. Key contributions include defining clusters for data aggregation, proposing a trust-based architecture, and applying belief functions within the proposed “Trust Region” to define boundaries and enable interpretable, reliable AI-driven decisions within cluster. Through extensive evaluations, we demonstrate the effectiveness of the proposed approach in V2X scenarios to improve system reliability within a cluster. Our findings form the basis for promoting trust in AI-based decision support for cluster-level V2X communication systems and pave the way for safer and more reliable autonomous transportation.